Importar librerías¶

  • Pandas para el uso de datos
  • NumPy para operaciones matemáticas
  • Scipy para ajuste de curvas
  • Matplotlib.pyplot para graficar
  • Sklearn.metrics para calcular métricas
  • SQLAlchemy para conexión a base de datos
In [1]:
import pandas as pd
import numpy as np
from scipy.optimize import curve_fit
from scipy.optimize import minimize
from scipy.optimize import least_squares
import matplotlib.pyplot as plt
from sklearn.metrics import mean_squared_error, r2_score, mean_absolute_error
import scipy 
from sqlalchemy import create_engine, inspect
from sqlalchemy.orm import sessionmaker
import datetime
import warnings
warnings.filterwarnings('ignore')

Conexión a base de datos¶

Para la base de datos se utiliza el servicio de https://neon.tech/, el cual se utiliza por integrar PostGIS sin costo y ofrecer una capacidad suficiente para pruebas.

In [2]:
db_url = "postgresql://test_owner:lpPrcHty8i9V@ep-royal-glitter-a50h4o6a.us-east-2.aws.neon.tech/test?sslmode=require"
engine = create_engine(db_url)

Lectura de Base de datos¶

Se guardó toda la información de la Especie con identificador 1 para hacer pruebas.

In [3]:
query = "SELECT * FROM data_especie1;"
df = pd.read_sql_query(query, engine)
df
Out[3]:
PARCELA # ROTACIÓN # ZONA AÑO PLANTACIÓN EDAD MEDICIÓN # FECHA MEDICIÓN ESPECIE PROCEDENCIA GENÉTICA ARBOL # ... DAP MEDIO [cm] AREA BASAL [m2/ha] H TOTAL [m] H DOMINANTE [m] H PODA [m] N/ha PODADOS [#] VOLUMEN EN PIÉ d.u. = 5 cm. [m3/ha scc]. IMA s/raleo [m3/ha/año]. Observación Clasificacion Densidad
0 1 1 1004 2012 6.043836 2 2018-06-20 00:00:00 1 Livingston HSCA T416 1 ... 17.229788 14.112918 9.768085 10.925 3.555319 None 51.132477 8.460269 None 3
1 1 1 1004 2012 6.043836 2 2018-06-20 00:00:00 1 Livingston HSCA T416 2 ... 17.229788 14.112918 9.768085 10.925 3.555319 None 51.132477 8.460269 None 3
2 1 1 1004 2012 6.043836 2 2018-06-20 00:00:00 1 Livingston HSCA T416 3 ... 17.229788 14.112918 9.768085 10.925 3.555319 None 51.132477 8.460269 None 3
3 1 1 1004 2012 6.043836 2 2018-06-20 00:00:00 1 Livingston HSCA T416 4 ... 17.229788 14.112918 9.768085 10.925 3.555319 None 51.132477 8.460269 None 3
4 1 1 1004 2012 6.043836 2 2018-06-20 00:00:00 1 Livingston HSCA T416 5 ... 17.229788 14.112918 9.768085 10.925 3.555319 None 51.132477 8.460269 None 3
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
383240 106 1 1004 2012 10.909589 7 2023-07-10 00:00:00 1 HSCB R 35 ... 28.300000 25.390000 17.600000 18.000 4.900000 None 173.100000 15.866775 None 2
383241 106 1 1004 2012 10.909589 7 2023-07-10 00:00:00 1 HSCB R 36 ... 28.300000 25.390000 17.600000 18.000 4.900000 None 173.100000 15.866775 None 2
383242 106 1 1004 2012 10.909589 7 2023-07-10 00:00:00 1 HSCB R 37 ... 28.300000 25.390000 17.600000 18.000 4.900000 None 173.100000 15.866775 None 2
383243 106 1 1004 2012 10.909589 7 2023-07-10 00:00:00 1 HSCB R 38 ... 28.300000 25.390000 17.600000 18.000 4.900000 None 173.100000 15.866775 None 2
383244 106 1 1004 2012 10.909589 7 2023-07-10 00:00:00 1 HSCB R 39 ... 28.300000 25.390000 17.600000 18.000 4.900000 None 173.100000 15.866775 None 2

383245 rows × 42 columns

Definición de modelos¶

Para el crecimiento se definieron 12 modelos de crecimiento. El ejercicio consiste en encontrar, a partir de los datos, los mejores valores constantes para cada modelo.

In [4]:
#Sweda
#N(t) = N0 * e^(r(t-t0))
def SWEDA(x,N0,r,t0):
    return N0 * np.exp(r*(x - t0))
    #return x * N0
#La fórmula para el modelo de crecimiento de Gompertz es la siguiente:
#N(t) = N0 * e^(-a * e^(-bt))
def GOMPERTZ(x,N0,a,b):
    return N0 * np.exp(-a * np.exp(-b * x))

#La función logística se puede escribir de la siguiente manera:#
#N(t) = K / (1 + ae^(-rt))
def LOGISTICA(x,k,a,r):
    return k / (1 + a * np.exp(-r * x))

def SIGMOID(x, L, k ,x0):
    """
    Modelo sigmoidal (función logística).

    Parámetros:
    - x: Array de valores independientes.
    - L: Límite superior de la curva.
    - x0: Punto medio de la curva en el eje x.
    - k: Pendiente de la curva en el punto medio.

    Retorna:
    - Array con los valores de la función sigmoidal para cada valor de x.
    """
    return L / (1 + np.exp(-k * (x - x0)))
def GOMPERTZ(x,N0,a,b):
    return N0 * np.exp(-a * np.exp(-b * x))

def GOMPERTZ2(x,N0,p,b):
    return N0 * np.exp(- np.exp(p -b * x))

def relacionPolimorfica(x,a,b,c):
    return a * ( 1 - np.exp( -b * x)) ** c

def MITSCHERLICH(x,a,L,b):
    return a * ( 1 - L* np.exp( -b * x))

#
def ChapmanRichards(x,a,b,c):
    return a * (1 -np.exp( -b * x)) ** c + np.exp(1)
                
def HossfeldI(x,a,b,c):
    return ((x ** 2) / (a + b*x) ** 2) + np.exp(1)
                
def Schumacher(x,a,b,c):
    return a * np.exp(-b * ( 1/ x)) + np.exp(1)
                
def Weibull(x,a,b,c):
    return a * ( 1 - np.exp( -b * x ** c)) + np.exp(1)

Funciones para modelar¶

En esta sección se definen un conjunto de funciones para encontrar el mejor modelo:

  • evaluar_modelo: Genera diferentes métricas para los valores encontrados.
  • getMinimosCuadrados: A través de la técnica de mínimos cuadrados de manera iterativa se encuentran los mejores valores para ajustar los modelos de la mejor manera.
  • getGrafico2: Con esta función se grafican los datos y la curva ajustada.
  • MODELOS2: Se evalúan todos los modelos definidos anteriormente.
In [5]:
def evaluar_modelo(funcion, X, Y,popt):  
    predicciones = funcion(X, *popt)
    # Calcular métricas básicas
    mse = mean_squared_error(Y, predicciones)
    r2 = r2_score(Y, predicciones)
    mae = mean_absolute_error(Y, predicciones)
    mape = np.mean(np.abs((Y - predicciones) / Y)) * 100
    std_residuos = np.std(Y - predicciones)
    
    # Otras métricas adicionales
    suma_residuos_abs = np.sum(np.abs(Y - predicciones))
    max_error_abs = np.max(np.abs(Y - predicciones))
    mean_error_abs = np.mean(np.abs(Y - predicciones))
    
    # Métricas adicionales
    suma_residuos_cuad = np.sum((Y - predicciones)**2)
    mediana_residuos = np.median(Y - predicciones)
    percentil_90 = np.percentile(Y - predicciones, 90)
    varianza_residuos = np.var(Y - predicciones)
    suma_cuad_residual_abs = np.sum(np.abs((Y - predicciones)) ** 2)
    min_error_abs = np.min(np.abs(Y - predicciones))
    skew_residuos = scipy.stats.skew(Y - predicciones)
    kurtosis_residuos = scipy.stats.kurtosis(Y - predicciones)
    coef_variacion_residuos = np.std(Y - predicciones) / np.mean(Y - predicciones)
    error_cuadratico_medio = np.sqrt(np.mean((Y - predicciones) ** 2))
    
    metricas = {
        "MSE": mse,
        "R2": r2,
        "MAE": mae,
        "MAPE": mape,
        "STD Residuos": std_residuos,
        "Suma Residuos Abs": suma_residuos_abs,
        "Max Error Abs": max_error_abs,
        "Mean Error Abs": mean_error_abs,
        "Suma Residuos Cuadráticos": suma_residuos_cuad,
        "Mediana Residuos": mediana_residuos,
        "Percentil 90": percentil_90,
        "Varianza Residuos": varianza_residuos,
        "Suma Cuadrada de los Residuos Absolutos": suma_cuad_residual_abs,
        "Min Error Abs": min_error_abs,
        "Skewness de los Residuos": skew_residuos,
        "Kurtosis de los Residuos": kurtosis_residuos,
        "Coeficiente de Variación de los Residuos": coef_variacion_residuos,
        "Error Cuadrático Medio": error_cuadratico_medio
    }

    return metricas


def getMinimosCuadrados(funcion,X,Y):
    def residuals(params, x, y):
        return y - funcion(x, *params)
    initial_guess = [1,1,1]
    result = least_squares(residuals, initial_guess, args=(X, Y))
    N0, t0, r = result.x
    return result.x

# Establecer la visualización en línea y mejorar la calidad de la imagen
%matplotlib inline
%config InlineBackend.figure_format = 'retina'

def getGrafico2(funcion, X, Y):
    # Generar puntos para el ajuste
    t_fit = np.linspace(min(X), max(X), 100) 
    
    # Obtener los parámetros del ajuste
    a,b,c = getMinimosCuadrados(funcion,X,Y)
    
    print(funcion.__name__,a,b,c)
    # Evaluar la función ajustada en los puntos de ajuste
    metricas = evaluar_modelo(funcion, X, Y,[a,b,c])
    N_fit = funcion(t_fit,a,b,c)
    
    # Plotear los datos y el ajuste
    plt.figure(figsize=(15, 9))  # Tamaño de la figura
    plt.scatter(X, Y, color='blue', label='Datos')  # Plotear los datos
    plt.plot(t_fit, N_fit, 'r-', label=f'Ajuste {funcion.__name__}')  # Plotear el ajuste
    plt.xlabel('Tiempo [Años]')
    plt.ylabel('DAP [mm]')
    plt.legend()
    plt.grid(True)  # Agregar una rejilla
    plt.title('Gráfico de Ajuste')  # Título del gráfico
    plt.tight_layout()  # Ajustar el diseño de la figura para evitar superposiciones
    plt.show()
    
    # Obtener métricas del modelo
    #metricas = evaluar_modelo(funcion, X, Y, params)
    metricas["Funcion"] = funcion.__name__
    metricas["b0"] = a
    metricas["b1"] = b
    metricas["b2"] = c
    return metricas

def MODELOS2(X,Y):
    d1 = getGrafico2(SWEDA,X,Y)
    d2 = getGrafico2(GOMPERTZ,X,Y)
    d3 = getGrafico2(LOGISTICA,X,Y)
    d4 = getGrafico2(SIGMOID,X,Y)
    d5 = getGrafico2(GOMPERTZ2,X,Y)
    d6 = getGrafico2(relacionPolimorfica,X,Y)
    d7 = getGrafico2(MITSCHERLICH,X,Y)
    d8 = getGrafico2(ChapmanRichards,X,Y)
    d9 = getGrafico2(HossfeldI,X,Y)
    d10 =getGrafico2(Schumacher,X,Y)
    d11 =getGrafico2(Weibull,X,Y)
    return  [d1,d2,d3,d4,d5,d6,d7,d8,d9,d10,d11]
In [6]:
dfClean = df[['Clasificacion Densidad','EDAD','DAP [mm]']].dropna()

Ejecución de todos los modelos¶

Esta es la parte del proceso más larga y que requiere mayor capacidad de cómputo.

In [10]:
acumulador = []
for i in dfClean['Clasificacion Densidad'].sort_values().unique():
    print(i)
    aux2 = dfClean[dfClean['Clasificacion Densidad'] == i] 
    auxX = np.array(aux2['EDAD'])
    auxY = np.array(aux2['DAP [mm]'])
    
    data = pd.DataFrame(MODELOS2(auxX,auxY))
    data["Clasificacion Densidad"] = i
    acumulador.append(data.copy())
errores = pd.concat(acumulador)
errores["Fecha_Analisis"] = datetime.datetime.now()

1
SWEDA 21.04119438647293 0.046419999139448886 7.10587799699723

GOMPERTZ 38.46327179557505 1.725621174054583 0.13785611393744854

LOGISTICA 37.032921601250976 3.2109256850486454 0.18705747048150667

SIGMOID 37.03315033104063 0.18705474512879688 6.236430348041789

GOMPERTZ2 38.46292029061756 0.5455947007342655 0.13785926728915074

relacionPolimorfica 44.160254594609384 0.0685756872552295 0.8134078716935539

MITSCHERLICH 41.366632283607444 0.9515908033103654 0.08787302045324784

ChapmanRichards 40.27005479755286 0.07826878248230645 0.9718266320152252

HossfeldI -0.6570581261937979 -0.1451961919079237 1.0

Schumacher 41.20107176680019 5.904595978915142 1.0

Weibull 41.09688061447227 0.08410061768487219 0.9664273943593613

2
SWEDA 26.7977474340287 0.04110951259995371 13.048828986525129

GOMPERTZ 32.847986171835934 1.869889046079423 0.19266388154042485

LOGISTICA 32.141472777929074 3.3644651051822088 0.24631612894827698

SIGMOID 32.141771309157946 0.24630685128190405 4.925693583241155

GOMPERTZ2 32.848188294752504 0.6258605254777986 0.19265915728445182

relacionPolimorfica 34.28552153982284 0.136328592007583 1.0709550122297369

MITSCHERLICH 34.097155236975155 1.0558302640322947 0.13786677039164807

ChapmanRichards 31.26552742874629 0.1463990049759853 1.2905145343274544

HossfeldI -0.5817228932128627 -0.15437065403552702 1.0

Schumacher 37.96821614948784 5.278386914371598 1.0

Weibull 31.173854585170137 0.0908874727180461 1.1377413194383494

3
SWEDA 7.904541477173488 0.046135017604025 -12.430984779173226

GOMPERTZ 27.417885945664782 2.7638464263107916 0.32492181842807744

LOGISTICA 27.010507377448942 5.559957955963131 0.4122874924427619

SIGMOID 27.010719004217545 0.41227226785021376 4.161162055555551

GOMPERTZ2 27.417877119669665 1.016625096151718 0.32492228697995723

relacionPolimorfica 27.866471749517146 0.27097624749715055 1.8579808791121428

MITSCHERLICH 28.155896614641886 1.3943535166740737 0.23578544655074993

ChapmanRichards 24.995769323947336 0.2891011771224708 2.361427329151902

HossfeldI 0.5270164563515811 0.16431834813746068 1.0

Schumacher 33.74423906381523 4.552423272287056 1.0

Weibull 24.732302633914294 0.06866740552596795 1.5004337935160004

4
SWEDA 10.948717094242452 0.04683918531239231 -2.1448902246631403

GOMPERTZ 24.63496096817586 2.5143897687622228 0.3118923717541098

LOGISTICA 24.302494065552892 5.435856801380573 0.40875746568228505

SIGMOID 24.302477433229928 0.40876076941455314 4.14186312703919

GOMPERTZ2 24.63492292473125 0.9220467561699824 0.3118966657815703

relacionPolimorfica 24.94745344909214 0.2571809181970153 1.633911218260609

MITSCHERLICH 25.200635532325702 1.266753195231155 0.22489617520766764

ChapmanRichards 22.08005829602988 0.2800462415621691 2.186642932784332

HossfeldI -0.5454844361071061 -0.17891782223103614 1.0

Schumacher 28.663112017828087 4.303202211625802 1.0

Weibull 21.702974607934003 0.06896985555569726 1.5081035460834373

5
SWEDA 18.23370649645544 0.013214082468090027 -2.934876255408325

GOMPERTZ 235.92070008524024 2.528636175352281 0.005647884883983733

LOGISTICA 118325.71013388612 6241.5346777196 0.013216350853996296

SIGMOID 699.614709769813 0.013658047404298507 262.2294850829726

GOMPERTZ2 164.6892201707684 0.7749999929759537 0.006672715370908827

relacionPolimorfica 72.08409580568882 0.00015224729655643197 0.18693111266696805

MITSCHERLICH 150.6947713938416 0.8769853033387759 0.0023475514379820784

ChapmanRichards 72.0159104663412 0.0001759500464682113 0.2123818143589132

HossfeldI -0.3328710561882554 -0.19915040073010296 1.0

Schumacher 24.888062914706094 2.9745310123974367 1.0

Weibull 83.2311987722169 0.14516978472886896 0.24416362612564754
In [11]:
errores
Out[11]:
MSE R2 MAE MAPE STD Residuos Suma Residuos Abs Max Error Abs Mean Error Abs Suma Residuos Cuadráticos Mediana Residuos ... Skewness de los Residuos Kurtosis de los Residuos Coeficiente de Variación de los Residuos Error Cuadrático Medio Funcion b0 b1 b2 Clasificacion Densidad Fecha_Analisis
0 27.558134 0.468702 4.029327 17.917199 5.249462 12434.504221 25.503244 4.029327 8.504440e+04 -0.300545 ... 0.368386 1.467820 -1.467915e+02 5.249584 SWEDA 21.041194 0.046420 7.105878 1 2024-05-02 17:10:38.800778
1 24.920378 0.519556 3.803969 16.033188 4.992028 11739.048860 24.255991 3.803969 7.690429e+04 -0.309014 ... 0.386523 1.732474 -8.306736e+02 4.992031 GOMPERTZ 38.463272 1.725621 0.137856 1 2024-05-02 17:10:38.800778
2 25.099417 0.516104 3.824390 16.227436 5.009920 11802.066760 24.268368 3.824390 7.745680e+04 -0.294159 ... 0.381045 1.697224 -4.531408e+02 5.009932 LOGISTICA 37.032922 3.210926 0.187057 1 2024-05-02 17:10:38.800778
3 25.099417 0.516104 3.824391 16.227448 5.009920 11802.069612 24.268388 3.824391 7.745680e+04 -0.294147 ... 0.381043 1.697223 -4.531715e+02 5.009932 SIGMOID 37.033150 0.187055 6.236430 1 2024-05-02 17:10:38.800778
4 24.920378 0.519556 3.803968 16.033172 4.992028 11739.045338 24.255969 3.803968 7.690429e+04 -0.308969 ... 0.386525 1.732475 -8.306188e+02 4.992031 GOMPERTZ2 38.462920 0.545595 0.137859 1 2024-05-02 17:10:38.800778
5 24.692280 0.523954 3.777398 15.808899 4.969129 11657.049170 24.326984 3.777398 7.620038e+04 -0.262152 ... 0.398153 1.787827 -8.413321e+02 4.969133 relacionPolimorfica 44.160255 0.068576 0.813408 1 2024-05-02 17:10:38.800778
6 24.727972 0.523266 3.779970 15.795186 4.972723 11664.988869 24.263190 3.779970 7.631052e+04 -0.236278 ... 0.391794 1.774221 7.188269e+06 4.972723 MITSCHERLICH 41.366632 0.951591 0.087873 1 2024-05-02 17:10:38.800778
7 24.730752 0.523212 3.782651 15.864034 4.972995 11673.260247 24.324262 3.782651 7.631910e+04 -0.287542 ... 0.397526 1.779102 -5.878602e+02 4.973002 ChapmanRichards 40.270055 0.078269 0.971827 1 2024-05-02 17:10:38.800778
8 24.684556 0.524103 3.771243 15.669904 4.968355 11638.056309 24.195639 3.771243 7.617654e+04 -0.198491 ... 0.401314 1.786289 2.991313e+03 4.968355 HossfeldI -0.657058 -0.145196 1.000000 1 2024-05-02 17:10:38.800778
9 24.873829 0.520454 3.782467 15.595337 4.987348 11672.694631 24.035584 3.782467 7.676064e+04 -0.173106 ... 0.413218 1.758948 3.638694e+02 4.987367 Schumacher 41.201072 5.904596 1.000000 1 2024-05-02 17:10:38.800778
10 24.728393 0.523257 3.782949 15.876494 4.972754 11674.181161 24.340027 3.782949 7.631182e+04 -0.284092 ... 0.399331 1.781259 -4.749267e+02 4.972765 Weibull 41.096881 0.084101 0.966427 1 2024-05-02 17:10:38.800778
0 20.924974 0.487241 3.592854 16.304069 4.574298 718797.055767 26.633854 3.592854 4.186313e+06 -0.081531 ... 0.051921 0.441579 -1.643146e+02 4.574382 SWEDA 26.797747 0.041110 13.048829 2 2024-05-02 17:10:38.800778
1 17.754864 0.564924 3.274430 14.064290 4.213650 655092.213581 29.536489 3.274430 3.552091e+06 -0.091188 ... 0.166082 0.705193 -1.045561e+03 4.213652 GOMPERTZ 32.847986 1.869889 0.192664 2 2024-05-02 17:10:38.800778
2 17.891695 0.561571 3.290563 14.190144 4.229850 658319.996740 29.685534 3.290563 3.579466e+06 -0.107287 ... 0.171539 0.688954 -5.148588e+02 4.229858 LOGISTICA 32.141473 3.364465 0.246316 2 2024-05-02 17:10:38.800778
3 17.891695 0.561571 3.290564 14.190167 4.229850 658320.199993 29.685372 3.290564 3.579466e+06 -0.107241 ... 0.171530 0.688944 -5.149224e+02 4.229858 SIGMOID 32.141771 0.246307 4.925694 2 2024-05-02 17:10:38.800778
4 17.754864 0.564924 3.274430 14.064303 4.213650 655092.312741 29.536401 3.274430 3.552091e+06 -0.091188 ... 0.166078 0.705188 -1.045660e+03 4.213652 GOMPERTZ2 32.848188 0.625861 0.192659 2 2024-05-02 17:10:38.800778
5 17.637967 0.567788 3.260695 13.971160 4.199756 652344.418822 29.286243 3.260695 3.528705e+06 -0.085934 ... 0.160186 0.717183 -1.107633e+03 4.199758 relacionPolimorfica 34.285522 0.136329 1.070955 2 2024-05-02 17:10:38.800778
6 17.627509 0.568044 3.259151 13.940096 4.198513 652035.611462 29.352917 3.259151 3.526612e+06 -0.072926 ... 0.159743 0.721301 4.776386e+07 4.198513 MITSCHERLICH 34.097155 1.055830 0.137867 2 2024-05-02 17:10:38.800778
7 17.662900 0.567177 3.263816 13.999101 4.202722 652968.891346 29.320816 3.263816 3.533693e+06 -0.093254 ... 0.161616 0.713528 -7.593367e+02 4.202725 ChapmanRichards 31.265527 0.146399 1.290515 2 2024-05-02 17:10:38.800778
8 17.618006 0.568277 3.259083 14.008167 4.197362 652022.015258 28.857013 3.259083 3.524711e+06 -0.110547 ... 0.154491 0.712372 -3.361600e+02 4.197381 HossfeldI -0.581723 -0.154371 1.000000 2 2024-05-02 17:10:38.800778
9 17.558918 0.569725 3.251401 13.868991 4.190336 650485.088840 29.338565 3.251401 3.512890e+06 -0.060965 ... 0.163964 0.734157 5.770884e+03 4.190336 Schumacher 37.968216 5.278387 1.000000 2 2024-05-02 17:10:38.800778
10 17.679527 0.566770 3.266090 14.029858 4.204694 653423.704274 29.272784 3.266090 3.537019e+06 -0.100046 ... 0.161674 0.709074 -4.867760e+02 4.204703 Weibull 31.173855 0.090887 1.137741 2 2024-05-02 17:10:38.800778
0 18.027486 0.431650 3.313443 20.391162 4.245769 354114.328696 24.570367 3.313443 1.926633e+06 0.094131 ... -0.182258 0.507175 -1.390072e+02 4.245879 SWEDA 7.904541 0.046135 -12.430985 3 2024-05-02 17:10:38.800778
1 13.464436 0.575508 2.843233 15.995061 3.669387 303861.962689 24.473881 2.843233 1.438971e+06 0.061531 ... -0.062884 0.770560 -6.198001e+02 3.669392 GOMPERTZ 27.417886 2.763846 0.324922 3 2024-05-02 17:10:38.800778
2 13.630683 0.570267 2.863779 16.227922 3.691956 306057.764521 23.966998 2.863779 1.456738e+06 0.043996 ... -0.053009 0.737879 -3.070714e+02 3.691976 LOGISTICA 27.010507 5.559958 0.412287 3 2024-05-02 17:10:38.800778
3 13.630683 0.570267 2.863780 16.227966 3.691956 306057.864109 23.966852 2.863780 1.456738e+06 0.043962 ... -0.053019 0.737873 -3.071021e+02 3.691976 SIGMOID 27.010719 0.412272 4.161162 3 2024-05-02 17:10:38.800778
4 13.464436 0.575508 2.843233 15.995059 3.669387 303861.960847 24.473886 2.843233 1.438971e+06 0.061530 ... -0.062883 0.770560 -6.198033e+02 3.669392 GOMPERTZ2 27.417877 1.016625 0.324922 3 2024-05-02 17:10:38.800778
5 13.381291 0.578130 2.833141 15.907482 3.658040 302783.494350 24.644897 2.833141 1.430085e+06 0.056667 ... -0.068576 0.786023 -6.372605e+02 3.658045 relacionPolimorfica 27.866472 0.270976 1.857981 3 2024-05-02 17:10:38.800778
6 13.315753 0.580196 2.825794 15.790662 3.649076 301998.256407 25.033095 2.825794 1.423081e+06 0.066861 ... -0.072811 0.800751 7.850994e+09 3.649076 MITSCHERLICH 28.155897 1.394354 0.235785 3 2024-05-02 17:10:38.800778
7 13.420908 0.576881 2.837977 15.970967 3.663444 303300.271140 24.482108 2.837977 1.434319e+06 0.054823 ... -0.065801 0.776998 -3.975777e+02 3.663456 ChapmanRichards 24.995769 0.289101 2.361427 3 2024-05-02 17:10:38.800778
8 13.812197 0.564544 2.886365 16.708567 3.716255 308471.565983 23.628713 2.886365 1.476137e+06 -0.015858 ... -0.077432 0.731722 -9.168369e+01 3.716476 HossfeldI 0.527016 0.164318 1.000000 3 2024-05-02 17:10:38.800778
9 13.441056 0.576245 2.839517 16.132440 3.666146 303464.907904 23.920111 2.839517 1.436473e+06 0.015731 ... -0.083088 0.789841 -1.763321e+02 3.666205 Schumacher 33.744239 4.552423 1.000000 3 2024-05-02 17:10:38.800778
10 13.511164 0.574035 2.848940 16.137520 3.675713 304471.949697 24.046107 2.848940 1.443965e+06 0.037299 ... -0.062217 0.760439 -2.137498e+02 3.675754 Weibull 24.732303 0.068667 1.500434 3 2024-05-02 17:10:38.800778
0 22.768430 0.407441 3.717773 21.285520 4.771564 36151.621336 21.742487 3.717773 2.214002e+05 0.258657 ... -0.317167 0.598934 -1.939608e+02 4.771628 SWEDA 10.948717 0.046839 -2.144890 4 2024-05-02 17:10:38.800778
1 19.708693 0.487072 3.314354 17.742434 4.439447 32228.780910 25.748176 3.314354 1.916473e+05 0.168855 ... -0.380165 1.316759 -1.192095e+04 4.439447 GOMPERTZ 24.634961 2.514390 0.311892 4 2024-05-02 17:10:38.800778
2 19.728109 0.486566 3.316106 17.751314 4.441633 32245.809967 25.716544 3.316106 1.918361e+05 0.142441 ... -0.372837 1.312863 -2.928909e+03 4.441634 LOGISTICA 24.302494 5.435857 0.408757 4 2024-05-02 17:10:38.800778
3 19.728109 0.486566 3.316106 17.751312 4.441633 32245.813457 25.716568 3.316106 1.918361e+05 0.142413 ... -0.372836 1.312863 -2.929713e+03 4.441634 SIGMOID 24.302477 0.408761 4.141863 4 2024-05-02 17:10:38.800778
4 19.708693 0.487072 3.314355 17.742433 4.439447 32228.788303 25.748212 3.314355 1.916473e+05 0.168858 ... -0.380163 1.316758 -1.193594e+04 4.439447 GOMPERTZ2 24.634923 0.922047 0.311897 4 2024-05-02 17:10:38.800778
5 19.702847 0.487224 3.314296 17.743217 4.438789 32228.212683 25.751296 3.314296 1.915905e+05 0.170144 ... -0.383236 1.316235 -2.660372e+04 4.438789 relacionPolimorfica 24.947453 0.257181 1.633911 4 2024-05-02 17:10:38.800778
6 19.702088 0.487244 3.314675 17.745090 4.438703 32231.900955 25.752987 3.314675 1.915831e+05 0.182895 ... -0.384442 1.314700 3.119356e+08 4.438703 MITSCHERLICH 25.200636 1.266753 0.224896 4 2024-05-02 17:10:38.800778
7 19.704387 0.487184 3.314147 17.742788 4.438962 32226.762725 25.748109 3.314147 1.916055e+05 0.163644 ... -0.382019 1.316798 -9.929461e+03 4.438962 ChapmanRichards 22.080058 0.280046 2.186643 4 2024-05-02 17:10:38.800778
8 19.971276 0.480238 3.353096 18.226032 4.468841 32605.502243 24.637879 3.353096 1.942007e+05 0.139372 ... -0.388635 1.232937 -1.650379e+02 4.468923 HossfeldI -0.545484 -0.178918 1.000000 4 2024-05-02 17:10:38.800778
9 19.768027 0.485528 3.320928 17.869996 4.446113 32292.707937 25.271326 3.320928 1.922243e+05 0.162979 ... -0.397631 1.299004 -4.300907e+02 4.446125 Schumacher 28.663112 4.303202 1.000000 4 2024-05-02 17:10:38.800778
10 19.725199 0.486642 3.315182 17.757861 4.441305 32236.830738 25.658915 3.315182 1.918078e+05 0.142658 ... -0.377916 1.314088 -1.232914e+03 4.441306 Weibull 21.702975 0.068970 1.508104 4 2024-05-02 17:10:38.800778
0 19.617978 0.008276 3.463919 17.266988 4.429219 4406.104867 20.011502 3.463919 2.495407e+04 0.135984 ... -0.344739 0.549655 3.378173e+04 4.429219 SWEDA 18.233706 0.013214 -2.934876 5 2024-05-02 17:10:38.800778
1 19.618835 0.008232 3.463991 17.266922 4.429315 4406.196400 20.010917 3.463991 2.495516e+04 0.135840 ... -0.344706 0.549592 6.360034e+03 4.429315 GOMPERTZ 235.920700 2.528636 0.005648 5 2024-05-02 17:10:38.800778
2 19.617978 0.008276 3.463919 17.266992 4.429219 4406.104457 20.011487 3.463919 2.495407e+04 0.135981 ... -0.344739 0.549655 3.564685e+04 4.429219 LOGISTICA 118325.710134 6241.534678 0.013216 5 2024-05-02 17:10:38.800778
3 19.618046 0.008272 3.463923 17.267025 4.429226 4406.109822 20.011394 3.463923 2.495415e+04 0.135903 ... -0.344736 0.549651 4.023631e+04 4.429226 SIGMOID 699.614710 0.013658 262.229485 5 2024-05-02 17:10:38.800778
4 19.618994 0.008224 3.464038 17.266181 4.429333 4406.256430 20.011946 3.464038 2.495536e+04 0.136995 ... -0.344700 0.549578 2.255960e+03 4.429333 GOMPERTZ2 164.689220 0.775000 0.006673 5 2024-05-02 17:10:38.800778
5 19.628327 0.007753 3.464777 17.270822 4.430387 4407.196797 20.009689 3.464777 2.496723e+04 0.123187 ... -0.344512 0.549678 6.286813e+03 4.430387 relacionPolimorfica 72.084096 0.000152 0.186931 5 2024-05-02 17:10:38.800778
6 19.620314 0.008158 3.464086 17.267345 4.429482 4406.317166 20.009021 3.464086 2.495704e+04 0.134767 ... -0.344650 0.549480 5.623966e+03 4.429482 MITSCHERLICH 150.694771 0.876985 0.002348 5 2024-05-02 17:10:38.800778
7 19.628069 0.007766 3.464760 17.270828 4.430358 4407.175108 20.009945 3.464760 2.496690e+04 0.123203 ... -0.344522 0.549701 7.630526e+03 4.430358 ChapmanRichards 72.015910 0.000176 0.212382 5 2024-05-02 17:10:38.800778
8 19.637312 0.007298 3.465495 17.274856 4.431401 4408.109610 20.007648 3.465495 2.497866e+04 0.117579 ... -0.344319 0.549698 5.590639e+04 4.431401 HossfeldI -0.332871 -0.199150 1.000000 5 2024-05-02 17:10:38.800778
9 19.638365 0.007245 3.465557 17.275199 4.431519 4408.188310 20.005824 3.465557 2.498000e+04 0.121227 ... -0.344272 0.549576 8.920082e+04 4.431519 Schumacher 24.888063 2.974531 1.000000 5 2024-05-02 17:10:38.800778
10 19.628426 0.007748 3.464769 17.271172 4.430398 4407.185879 20.009102 3.464769 2.496736e+04 0.122584 ... -0.344509 0.549674 2.255647e+04 4.430398 Weibull 83.231199 0.145170 0.244164 5 2024-05-02 17:10:38.800778

55 rows × 24 columns

Guardar la base de datos de errores y constantes¶

In [13]:
errores.to_sql("data_error", engine, if_exists='append', index=False)
Out[13]:
55
In [ ]: